From: | Peter Geoghegan <pg(at)bowt(dot)ie> |
---|---|
To: | Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com> |
Cc: | Bhushan Uparkar <bhushan(dot)uparkar(at)gmail(dot)com>, Jesper Pedersen <jesper(dot)pedersen(at)redhat(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, Andrew Dunstan <andrew(dot)dunstan(at)2ndquadrant(dot)com>, Alexander Korotkov <a(dot)korotkov(at)postgrespro(dot)ru> |
Subject: | Re: Index Skip Scan |
Date: | 2018-08-17 17:52:05 |
Message-ID: | CAH2-Wzkk0+vSAiaRSkX84KR0SofKkY1yGHEcLoRBXT-JjM37bw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Thu, Aug 16, 2018 at 4:10 PM, Thomas Munro
<thomas(dot)munro(at)enterprisedb(dot)com> wrote:
> Can you give an example of problematic ndistinct underestimation?
Yes. See https://postgr.es/m/CAKuK5J12QokFh88tQz-oJMSiBg2QyjM7K7HLnbYi3Ze+Y5BtWQ@mail.gmail.com,
for example. That's a complaint about an underestimation specifically.
This seems to come up about once every 3 years, at least from my
perspective. I'm always surprised that ndistinct doesn't get
implicated in bad query plans more frequently.
> I suppose you might be able to defend against that in the executor: if
> you find that you've done an unexpectedly high number of skips, you
> could fall back to regular next-tuple mode. Unfortunately that's
> require the parent plan node to tolerate non-unique results.
I like the idea of dynamic fallback in certain situations, but the
details always seem complicated.
--
Peter Geoghegan
From | Date | Subject | |
---|---|---|---|
Next Message | Alexander Korotkov | 2018-08-17 18:07:13 | Re: [HACKERS] WIP: long transactions on hot standby feedback replica / proof of concept |
Previous Message | Tom Lane | 2018-08-17 17:38:46 | Re: [HACKERS] WIP: long transactions on hot standby feedback replica / proof of concept |